Include usage guide

README.md

@@ -7,4 +7,59 @@ metrics:
 base_model:
 - facebook/mbart-large-50
 library_name: transformers
----
+---
+## How to use the model
+
+Import model and tokenizer using from stranformer libray
+```py
+# Load model directly
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+
+tokenizer = AutoTokenizer.from_pretrained("tykea/mBart-large-50-KQA")
+model = AutoModelForSeq2SeqLM.from_pretrained("tykea/mBart-large-50-KQA")
+```
+Define function to take question and pass to the model
+```py
+import torch
+
+#ask function for easier asking 
+def ask(custom_question):
+# Tokenize the input
+    inputs = tokenizer(
+        f"qestion: {custom_question}",
+        return_tensors="pt",
+        truncation=True,
+        max_length=512,
+        padding="max_length"
+    )
+
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    inputs = {key: value.to(device) for key, value in inputs.items()}
+
+    model.eval()
+    with torch.no_grad():
+        outputs = model.generate(
+        input_ids=inputs["input_ids"],
+        max_length=50,
+        num_beams=4, 
+        repetition_penalty=2.0,
+        early_stopping=True,
+        do_sample=True,     
+        top_k=50,           
+        top_p=0.95,         
+        temperature=0.7,
+    )
+    answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    print(f"Question: {custom_question}")
+    print(f"Answer: {answer}")
+
+```
+
+Then call the function #ask function 
+```py
+question = "តើរស់នៅប្រទេសណា"
+ask(question)
+#output
+#Question: តើរស់នៅប្រទេសណា
+#Answer: រស់នៅប្រទេសអង់គ្លេស
+```